DETECTION OF NARROW-BAND SIGNALS IN SPHERICALLY INVARIANT NOISE

This paper deals with noncoherent discrete-time detection of a narrow- band signal subject to slow and nonselective fading and embedded in co rrelated non-Gaussian noise modeled as a spherically invariant random process whose modulating random variable is continuous. At first, an a symptotic sufficient statistic for an arbitrary fading law is derived; then, the asymptotically optimum detector for Rayleigh-distributed am plitude fluctuations is synthesized. The detection structure implement ation requires the knowledge, but for a scale factor, of the correlati on function of the noise, but is independent of the distribution funct ion of the modulating random variable. The performance of the asymptot ically optimum detector synthesized for Rayleigh fading is assessed vi a computer simulations. The results show that the performance degradat ion with respect to the fully optimum performance is scarcely signific ant, even for low values of the sample size. Moreover, in highly non-G aussian noise the proposed detector largely outperforms the fully opti mum detector synthesized under the correlated Gaussian noise assumptio n. (C) 1997 Acoustical Society of America.